Tracking a Northern Fulmar from a Scottish Nesting Site to the Charlie

Home , Northern fulmar

Edwards, E. W.J., Quinn, L. R., Wakefield, E. D., Miller, P. I., and Thompson, P. M. (2013) Tracking a northern fulmar from a Scottish nesting site to the Charlie-Gibbs Fracture Zone: evidence of linkage between coastal breeding seabirds and Mid-Atlantic Ridge feeding sites. Deep-Sea Research Part II: Topical Studies in Oceanography, 98(B). pp. 438-444.

This work is made available under the Creative Commons Attribution NonCommercial-NoDerivs 3.0 Unported License (CC BY-NC-ND 3.0)

Version: Published

http://eprints.gla.ac.uk/103806/

Deposited on: 10 Mar 2015

Enlighten – Research publications by members of the University of Glasgow http://eprints.gla.ac.uk Deep-Sea Research II 98 (2013) 438–444

Contents lists available at ScienceDirect

Deep-Sea Research II

journal homepage: www.elsevier.com/locate/dsr2

Tracking a northern fulmar from a Scottish nesting site to the Charlie-Gibbs Fracture Zone: Evidence of linkage between coastal breeding seabirds and Mid-Atlantic Ridge feeding sites

Ewan W.J. Edwards a,n, Lucy R. Quinn a, Ewan D. Wakeﬁeld b, Peter I. Miller c, Paul M. Thompson a a University of Aberdeen, Institute of Biological and Environmental Sciences, Lighthouse Field Station, Cromarty, Ross-shire IV11 8YL, UK b Faculty of Biological Sciences, University of Leeds, Leeds LS2 9JY, UK c Plymouth Marine Laboratory, Prospect Place, The Hoe, Plymouth PL1 3DH, UK article info abstract

Available online 12 April 2013 The seas above mid-ocean ridges are biodiversity hotspots in an otherwise largely oligotrophic Keywords: environment, but the nature and extent of linkage between these offshore regimes and coastal Tracking ecosystems remains uncertain. Using a combination of GPS and geolocation tracking data, we show Marine birds that a male fulmar, breeding on the Scottish coast, foraged over areas of persistent thermal fronts along Mid-ocean ridges the Charlie-Gibbs Fracture Zone (CGFZ) of the Mid-Atlantic Ridge during the incubation period. The bird Oceanic fronts travelled over 6200 km in 14.9 days. First-passage time analysis identiﬁed seven areas of restricted Foraging behaviour search, four on the shelf and three in the vicinity of the Mid-Atlantic Ridge. Previous studies of incubation First-passage time foraging trip durations at this site suggest that a trip of this duration is unusual, and further work is Northern fulmar required to assess the extent to which different individuals use these offshore resources. Nevertheless, these data highlight the potential importance of high sea areas beyond the limits of national jurisdiction when considering the management and conservation of seabirds breeding in NW Europe, and raises the potential for even greater linkage between the CGFZ and seabirds breeding colonies in other regions. & 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license.

1. Introduction efforts to conserve their populations. For example, European legislation calls for specific action to protect birds and their habitats Inrecentyears,theseaabovetheCharlie-GibbsFractureZone (European Commission, 2009) with a particular focus on reduction (CGFZ) has become recognised as a region of rich biodiversity of incidental bycatch of seabirds in fishing gears within European (Letessier et al., 2011, 2012), providing an important foraging area Union waters (FAO, 2008) However outwith territorial waters the for a broad suite of marine predators, including fishes (Fossen et al., legislation is more difficult to enforce. 2008), seabirds (Bogdanova et al., 2011; Egevang et al., 2010; Previous surveys of seabirds at sea (e.g. Boertmann, 2011)have Frederiksen et al., 2012) and mammals (Doksæter et al., 2008; Skov been unable to determine the breeding origin, fidelity or frequency of et al., 2008). As a result, the region has been designated as a large High use of individuals observed in high seas areas such as the CGFZ. Seas Marine Protected Area by the Convention for the Protection of the Furthermore, seabirds do not recruit until they are several years old, Marine Environment of the North-East Atlantic (OSPAR Commission) and may skip breeding in some years. Consequently, even when and the North-East Atlantic Fisheries Commission (NEAFC) (OSPAR, seabirds have been observed in the CGFZ during the breeding season, 2010). However, the nature and extent of linkage between these these individuals may represent birds from the non-breeding compo- offshore areas and coastal ecosystems remains uncertain. Some of the nent of the population that spend extended periods foraging at sea. animals (e.g. cetaceans, fishes) feeding in these areas remain in the Tracking technologies have recently revealed that seabirds pelagic environment throughout their life. But seabirds must return to from Icelandic, Scottish and Canadian breeding colonies may land to breed, and an understanding of the relationship between these spend some of their time over the Mid-Atlantic Ridge (MAR) area high seas regions and seabird breeding sites is essential to support (Bogdanova et al., 2011; Egevang et al., 2010; Mallory et al., 2008a). Most published studies provide examples of seabirds foraging in this region during the winter, when individuals do not need to return regularly to coastal breeding colonies. To date, the only example of actively breeding adult birds that foraged over the CGFZ has been a study of Cory's shearwaters (Calonectris diome- n Corresponding author. dea), which breed approx. 1200 km away on the Azores archipe- E-mail address: [email protected] (E.W.J. Edwards). lago (Magalhães et al., 2008).

0967-0645 & 2013 The Authors. Published by Elsevier Ltd. Open access under CC BY-NC-ND license. http://dx.doi.org/10.1016/j.dsr2.2013.04.011 E.W.J. Edwards et al. / Deep-Sea Research II 98 (2013) 438–444 439

In this paper, we present tracking data from an adult male site, with the same partner, since 2001. This bird was captured on northern fulmar (Fulmarus glacialis) breeding on a Scottish island, the nest at 12:06 BST on May 23, 2012, while its female partner which demonstrate that seabirds breeding at colonies on the was on a foraging trip. Following capture, we removed a GLS European continental shelf can actively forage in the MAR area logger that had previously been ﬁtted to this male in July 2010, during the breeding season. This ﬁnding emphasises the trophic and a new GLS logger and GPS tag were then applied before he importance of high seas beyond the limits of national jurisdiction was resettled onto the nest. Both the new GLS logger and GPS tag when considering the management and conservation of seabirds, were then recovered at 10.00 BST on June 10, 2012, after the male and highlights the potential for even greater linkage between the bird had completed one foraging trip and returned to relieve the CGFZ and seabird breeding colonies in other regions. incubating female. Following recovery, the male bird was resettled on the nest.

2. Study species 3.2. Analyses

Northern fulmars, a circumpolar boreal species, are the most GPS data were downloaded using the manufacturer's software, widespread and abundant seabird in the North Atlantic, breeding and tracks plotted using ESRI Arc GIS 10. Great circle distances on coastal cliffs and grassy slopes from Arctic islands to the coast between each at-sea location and the nest site were calculated. of Brittany, France (Mitchell et al., 2004). As generalist predators Flight speeds between successive GPS locations were also calcu- fi and scavengers, they consume a variety of sh, squid, crustaceans lated using great circle distances. We identified core foraging or and offal (Phillips et al., 1999). Typical of other Procellariids, they resting areas using first-passage time (FPT) analysis (Fauchald and are long-lived (Grosbois and Thompson, 2005), with a breeding Tveraa, 2003; Pinaud and Weimerskirch, 2007), using the ‘adeha- period that extends, at Scottish colonies, from laying a single egg bitatLT’ package (Calenge, 2006) in R 2.12.2 (R Development Core fl in mid-May, to edging in late August (Dunnet, 1991). Foraging Team, 2008). trips during the breeding season are considered to be constrained Data on light levels and activity were downloaded from the GLS in range and duration because both partners share incubation logger, extracted and filtered using the BASTrak software suite (Mallory et al., 2008b), and then continue to take turns brooding (Fox, 2010). Sunsets and sunrises were manually identified using fl and feeding the chick until edging (Hamer and Thompson, 1997; the programme TransEdit; the solar angle of elevation was set Phillips and Hamer, 2000). Foraging trips during the breeding as −3.51. All data from the equinoxes were excluded to avoid season have been reported to have a maximum duration of 18 days periods when latitude estimation was not possible, and noon and (Mallory et al., 2008b), and assumed to be limited to a maximum midnight positions were then visualised in ArcGIS. Activity data, range of 580 km (Thaxter et al., 2012) based upon published available as the number of wet samples within each 10-min period information on foraging trip durations during chick rearing through the deployment, were then used to classify the activity (Furness and Todd, 1984). within each 10-min sample as completely dry, completely wet or mixed. Subsequently, sequences of these records were analysed to define the duration of extended bouts of each of these activity 3. Methods category. Remote sensing data from microwave (AMSR-E, TMI, WindSAT) The study was conducted at a fulmar breeding colony on and infrared sea surface temperature (SST; AVHRR, MODIS) sen- Eynhallow, an uninhabited island in Orkney, off the north coast sors were used to locate persistent oceanic fronts. Merging these of Scotland (5918′N, 317′W). This colony has been the subject of data, seven-day composite front maps (Miller, 2009) were derived. detailed demographic studies since 1950 (Dunnet, 1991). On-going More detailed descriptions of these methods are presented in tracking studies at Eynhallow seek to define the range of foraging Miller et al. (2013). by adults during the breeding season, and broad-scale winter distribution patterns. This paper presents data from tracking records that shows how far and for how long one fulmar forages. 4. Results

3.1. Datalogger deployment and recovery 4.1. General foraging trip pattern and areas of restricted search

To attach and recover dataloggers, fulmars were caught under GPS loggers were recovered from 12 out of 22 tagged birds. In licence on the nest using a net or noose. In May 2012, 22 birds May and June 2012, GPS tracks indicated that 10 male and 2 female were fitted with a GPS logger (attached to mantle feathers using birds made foraging trips ranging from 4 to 15 days during strips of Tesas waterproof tape; MobileActions iGot-U GT-120, incubation. Most trips were within 100 km of the breeding colony, weight 18 g after modification) and geolocator (using a cable tie one male travelled to the eastern North Sea (approx. 800 km). Of around a Darvic leg ring; BAS Mk15, 3.6 g). Together the devices particular interest was male fulmar, ♯1568, who flew 2500 km weigh 3% of the birds' body mass. The GPS was set to record west to the MAR as described in detail below. The foraging position every hour for the duration of the battery life or foraging characteristics of the remaining 11 birds will be reported trip, whichever came first. Geolocators (GLS) logged light levels elsewhere. (Phillips et al., 2004), and whether the device was wet or dry (see Following tagging on May 23, bird ♯1568 remained on the egg Mackley et al., 2011), every 3 s, and stored summary data (max- until the return of his partner. At around 10:30 BST on May 26, imum light level and number of wet samples) every 10 min. 2012, he left the nest site and spent the next 48 h to the northwest Although these devices were used on deployments over single of Orkney, before embarking on a sustained flight (11 h) to the foraging trips in May 2012, they had previously been fitted to over middle of the Shetland-Faeroes Channel (Fig. 1A). After spending 100 birds between 2008 and 2011, allowing daily data collection approximately 18 h around the Wyville Thompson Ridge, he until recovery in subsequent breeding seasons. headed WSW into the mid-Atlantic Ocean, remained around the The data presented in this paper were collected from an adult CGFZ region for 2–3 days and then headed due east. After 19 h his male (sexed using morphometric measurements, Dunnet and route deviated to the ESE towards Ireland, reaching Galway Bay Anderson, 1961), fulmar (♯1568), which had bred at the same nest where he spent 8 h before turning north along the Irish coast. 440 E.W.J. Edwards et al. / Deep-Sea Research II 98 (2013) 438–444

Fig. 1. Maps showing the foraging trip GPS track of bird ♯1568 from Eynhallow to the Mid-Atlantic Ridge and back. (A) Shows periods of night (dark circles) and light (white circles) on a bathymetric chart (darker colours indicate deeper water). (B) shows the regions of ARS as identiﬁed by ﬁrst-passage time (warmer colours indicate higher FPT). ARS regions are circled and numbered. (C) Shows the GPS track overlaid on a composite front map showing the position and strength of thermal fronts, averaged between 31 May 2012 and 06 June 2012 (darker greys indicate stronger fronts in terms of gradient and persistence). ARS regions are circled and numbered. E.W.J. Edwards et al. / Deep-Sea Research II 98 (2013) 438–444 441

After 18 h he ﬂew to the west of Tiree and the Outer Hebrides, Galway Bay, Ireland; and one W of the island of Tiree, Scotland. headed east at the northern tip of the Isle of Lewis, and followed Comparing these areas of ARS with oceanic fronts present at this the coast around the NW tip of Scotland to return to Eynhallow at time (Fig. 1C) revealed that the bird's activity around the CGFZ was 21:00 on June 9, 2012. concentrated near the SPF. Based upon great circle distances between the hourly GPS locations, the total foraging trip length was 6219 km, which the 4.2. Year-round distribution bird completed in 14.9 days. Data from the GLS logger provided information on local dawn and dusk (Fig. 2), and show how the The geolocator deployed on this bird in July 2010 indicated bird's distance from the colony varied through this period during that, whilst dispersing widely across the North Atlantic, including different light and dark periods. FPT analysis identiﬁed seven time in the Labrador Strait and Norwegian Sea, the bird had clearly apparent regions of area-restricted search (ARS) during the trip visited the CGFZ during the previous 24 months (Fig. 3). Also of (Fig. 1B): one on the Wyville Thompson Ridge in the Faeroes- note is that relatively few geolocator positions occur between the Shetland channel (601N, 71W); one on the Rosemary Bank, off the MAR and the UK, suggesting (as with this study) rapid transits NW of Scotland; three within the CGFZ/SPF region; one in outer between the MAR and European shelf waters.

Fig. 2. Graph showing distance from nest site on Eynhallow through time. Black circles indicate the location of ARS regions, as identiﬁed by FPT analysis. Numbers correspond to numbers in Fig. 1b and c. Local daylight and darkness indicated by light and dark bars.

Fig. 3. Map showing North Atlantic region, with foraging trip from this study (dashed line) presented alongside twice-daily locations from preceding 2 years of geolocator tracking (double-smoothed; Phillips et al., 2004). 442 E.W.J. Edwards et al. / Deep-Sea Research II 98 (2013) 438–444

4.3. Activity and flying speed Rank Sum test, W¼522, po0.001, df¼1), but there was no significant difference between flight speeds by day (median¼4.08, There were two periods of ARS at the beginning and end of the range¼0.81–24.9, n¼78) or night (median¼7.33, range¼0.42– trip, and three whilst the bird was around the MAR (Figs. 1b and 23.5, n¼21) during periods of ARS (Wilcoxon Rank Sum test, 2). We explored wet/dry patterns and flight speed within each of W¼1008, p¼0.1067, df¼1; Fig. 4). these ARS, and during the major transits across the Atlantic. Flying Periods of ARS varied from o3to430 h, with three occurring speeds during the transits were similar on both outward and completely in local daylight, two in complete or pre-dominant return journeys. The distance between ARS 2 on the Porcupine darkness, and the two longest being primarily in daylight (Table 1). Bank and ARS 3 at the CGFZ was 1610 km, and was flown in 55 h, The shortest of these periods was the last ARS, to the west of Tiree giving an overall travel speed of 28.6 km h−1. The maximum (Fig. 1B), when the bird spent almost all of its time on the water in hourly speed was 63.3 km h−1. This outward journey was asso- the dark. In all other cases, activity in the ARS regions showed a ciated with strong E/SE winds associated with a deep depression prevalence of both mixed and wet behaviours (median 94%, range (971mb at 0000 UTC, 31 May 2012) in the central North Atlantic. 83–100%), with very little time spent dry (median 5.5%, range On the return trip, the distance between ARS 5 at the CGFZ and 0–17%) (Table 1). Conversely, during the long transits to the ARS 6 in Galway Bay was 2055 km, and was flown in 73 h, giving east/west, the majority of time was spent dry (median 43%, range an overall travel speed of 27.7 km h−1 and a maximum hourly 42–44%) or engaged in mixed wet/dry behaviour (median 43%, speed of 48.7 km h−1. This return journey was undertaken in range 38–47%), with minimal time spent on the water (median largely headwind conditions, with a deepening low-pressure area 14%, range 8–20%). Within ARS areas, hourly GPS positions to the SW of the UK resulting in a northeasterly airflow. indicated that the bird still searched extensively, with average Flight speeds were faster during these transits (median¼30.1, distance covered per h ranging between 3.5 and 9.5 km. The range¼0.57–63.3, n¼134) than during periods of ARS (med- longest periods of mixed activity occurred during the two long ian¼4.43, range¼0.42–24.9, n¼99; Wilcoxon Rank Sum test, ARS over the CGFZ (Table 2). Data from the long transits suggest W¼1362, po0.001, df¼1). During transits, flight speeds were that sustained periods of flight typically lasted about 30 min, faster during the day (median¼33.8, range¼3.41–63.3, n¼105) interspersed with periods of mixed activity or total immersion than at night (median¼16.7, range¼0.57–34.1, n¼29; Wilcoxon on the water.

Fig. 4. Boxplots showing (A) the difference in flight speed between transits and regions of ARS; (B) the difference in flight speed between periods of light and dark during transits; and (C) the difference in flight speed between periods of light and dark during ARS.

Table 1 Table showing details of activity in ARS regions and two major transits.

Start date and End date and Duration Duration % Time in local Mean speed Max speed % 3-s samples Proportion of time time (h) (days) darkness (km/h) (km/h) wet 10 min periods

Dry Mixed Wet

ARS 1 5/29/2012 3:18 5/29/2012 16.62 0.69 0.0 3.52 – 78.13 0.10 0.33 0.57 19:55 ARS 2 5/30/2012 3:54 5/30/2012 6.87 0.29 0.0 8.26 – 79.15 0.02 0.45 0.52 10:46 ARS 3 6/1/2012 17:45 6/1/2012 20:44 2.98 0.12 0.0 7.10 – 87.94 0.00 0.39 0.61 ARS 4 6/2/2012 6:20 6/3/2012 3:32 21.20 0.88 22.7 9.52 – 59.47 0.06 0.53 0.41 ARS 5 6/3/2012 13:14 6/4/2012 19:24 30.16 1.26 22.0 6.93 – 59.17 0.05 0.62 0.33 ARS 6 6/7/2012 20:47 6/8/2012 4:27 7.68 0.32 80.4 9.05 – 59.45 0.17 0.50 0.33 ARS 7 6/8/2012 22:48 6/9/2012 1:46 2.96 0.12 100.0 2.22 – 97.06 0.00 0.08 0.92

Westbound 5/30/2012 10:46 6/1/2012 17:45 54.99 2.29 20.8 28.58 63.30 33.03 0.42 0.38 0.20 Eastbound 6/4/2012 19:24 6/7/2012 20:47 73.38 3.06 25.2 27.66 48.66 21.84 0.44 0.47 0.08 E.W.J. Edwards et al. / Deep-Sea Research II 98 (2013) 438–444 443

Table 2 followed by brief periods on the water to capture and consume prey Table listing descriptive statistics of activity bouts during ARS and transits. (Table 2). The majority of these episodes occurred during daylight hours, which is consistent with the visual nature of the fulmars' Dry bout durations Mixed bout Wet bout durations fl (min) durations (min) (min) foraging, but these activity patterns could also be in uenced by the diurnal vertical migration of prey in the region (Sutton et al., 2008). Mean SE Max n Mean SE Max n Mean SE Max n Activity data from the GLS loggers indicate that the bird also engaged in mixed activity during the long transits, where short ARS 1 37 27 90 3 23 5 70 14 54 22 190 11 flights were interspersed with regular contact with sea water. This ARS 2 10 – 10 1 30 9 70 6 37 10 70 6 ARS 3 –––060 2080 2110 1101 mixed activity may be indicative of foraging whilst travelling. ARS 4 27 17 60 3 88 32 300 9 104 30 190 5 Flight speed during these phases of the trip was similar during the ARS 5 18 6 40 5 81 24 290 14 75 25 190 8 westbound and eastbound transits, despite different environmen- ARS 6 62 43 190 4 23 4 40 10 25 6 50 6 tal conditions and wind strengths. Given the regular bouts of wet ARS 7 –––0 –––0210 – 210 1 or mixed activity (Table 2), and the assumption that speed is Westbound 34 4 130 57 31 4 160 67 34 8 90 11 uniform in a straight line between hourly position fixes, these are Eastbound 38 7 240 37 25 3 110 49 54 21 280 12 likely minimum estimates of flying speeds. Furthermore the meandering path of Procellariids engaged in dynamic soaring (Pennycuick, 2002; Tickell, 2000) is likely to result in much faster 5. Discussion instantaneous travel speeds than those speeds calculated from distance between hourly positions. Seabirds often forage in high sea regions distant from their GPS data highlight that the bird did not embark on the west- breeding colonies during post-breeding dispersal or the winter bound transit of this foraging trip immediately after leaving the (Frederiksen et al., 2012; Mallory et al., 2008a). However, our study nest, instead, spending several days to the NW of Orkney (Fig. 1A). highlights the potential linkage between coastal and mid-ocean Whilst the decision to remain in this area may have been due to ecosystems within the fulmar breeding season. Information on the favourable local foraging conditions, this period coincided with a duration of foraging trips during incubation is available from only large and stable anticyclone in the region. Previous research has a small number of individual fulmars (Mallory et al., 2008b), highlighted the high energetic cost that this species incurs when suggesting that, whilst such a long trip may be unusual, it was flying in low wind speeds, and it is possible that the delayed probably not exceptional. For example, incubation trip durations of departure to foraging grounds on the MAR was a response to 16 days and 18 days have been recorded for fulmars breeding at insufficient winds for energetically efficient flight (Furness and sites in Alaska and arctic Canada (Mallory et al., 2008b). Bryant, 1996). On the return leg, the fastest flight speeds were No previously published tracking data during the incubation attained 100 km to the west of the Irish coast, as the bird flew period exist to compare foraging ranges of fulmars at this or other close to an area of low pressure centred over the British Isles, sites, but we recorded such an extensive range in only one of the where wind speeds were probably the highest encountered during 12 individuals tracked using GPS loggers during our study. Pre- the foraging trip. Inspection of the activity patterns and locations vious satellite tracking studies of fulmars in northern Europe and within the trip suggest that in the absence of such adverse wind North America have shown that birds may travel long distances on conditions, this bird could have completed the foraging trip in just foraging trips, the furthest recorded being 32,000 km in 239 days 11 or 12 days. over the winter period (Mallory et al., 2008a), however most of The observations detailed within this paper highlight how GPS these data were collected during the post-breeding period or from and GLS tracking can now be used to explore the extent to which failed breeders (Falk and Moeller, 1997; Hatch et al., 2010; Mallory other breeders from this colony use distant offshore resources. et al., 2008a). Longer-term GLS data collected from bird ♯1568 Previous work at this colony provided evidence that reproductive indicates that he returned to the CGFZ region in other seasons and rates (Thompson and Ollason, 2001) and adult survival (Grosbois in other years (Fig. 3). Whilst these GLS data lack the accuracy of and Thompson, 2005)wereinfluenced by the North Atlantic GPS tracks, they do provide excellent opportunities for broader Oscillation (NAO). At the time, it was suggested that effects on scale and longer term studies that can in future be used to assess reproduction and survival could result from known relationships the extent to which different individuals use the MAR and similar between the NAO and potential prey stocks. However, the dis- offshore areas over extended periods of time. Fulmars tracked covery that birds make long foraging trips during the incubation using satellite tags from the Canadian high Arctic (Devon Island) period highlights that changes in wind speed related to the NAO were found to migrate into the NW Atlantic between Canada and (Pirazzoli et al., 2010) could affect the cost of flying to different Iceland, and East to the CGFZ during the winter (Mallory et al., foraging grounds (Furness and Bryant, 1996). In the Southern 2008a). Together, these studies highlight that fulmars from both Ocean, wandering albatrosses breeding on Crozet Island have North American and European colonies may overlap in their usage shifted their at-sea distribution southwards and shown higher of the MAR during the non-breeding season. breeding success, corresponding with an increase in intensity and First-passage time analysis demonstrated that the bird in this polewards shift in prevailing winds (Weimerskirch et al., 2012). study exhibited area-restricted searching behaviour at seven stages Future tracking studies should examine how variation in wind during its trip, both on the shelf and in the mid-Atlantic. The fields may influence fulmar foraging trip characteristics in differ- occurrence of oceanic thermal fronts over this period suggested that ent parts of their foraging range. the bird associated with areas known to be productivity hotspots In Pacific fulmar colonies (Hatch et al., 2010), one negative (Letessier et al., 2012). Furthermore, this bird appeared to concen- consequence of using distant foraging areas during the breeding trate its foraging in the region of the SPF. Little is known of the diet of season is the risk of mortality from being caught on hooks in a fulmars outside the chick-rearing period (Owen et al., in press). wide range of different longline fisheries. In this study, the single However, it is clear that fulmars are primarily surface-feeding visual fulmar travelled through areas where the Spanish demersal long- predators or scavengers (Garthe and Furness, 2001). In this study, the line fishery catches over 50,000 birds per year, mostly great small percentage of flight time and the high proportion of mixed shearwaters Puffinus gravis and fulmars (Anderson et al., 2011). activity during the longer periods of ARS suggest that foraging On the same trip, the bird traversed areas where Faroese demersal probably involves short searching flights within prey patches, longliners regularly record fulmars as bycatch. The possibility of 444 E.W.J. Edwards et al. / Deep-Sea Research II 98 (2013) 438–444 bycatch also exists from pelagic fisheries over the MAR (OSPAR, Goutte, M.L., Grémillet, D., Guilford, T., Jensen, G.H., Krasnov, Y., Lorentsen, S.-H., 2010). This study shows that even birds breeding several thousand Mallory, M.L., Newell, M., Olsen, B., Shaw, D., Steen, H., Strøm, H., Systad, G.H., Thórarinsson, T.L., Anker-Nilssen, T., 2012. Multicolony tracking reveals the kilometres from such areas may be at risk of bycatch. Many winter distribution of a pelagic seabird on an ocean basin scale. Diversity breeding colonies are situated outwith regions where longline Distrib. 18, 530–542. fishing occurs, but evidence is mounting that these birds are still at Furness, R.W., Bryant, D.M., 1996. Effect of wind on field metabolic rates of breeding – risk from fisheries bycatch. Consequently, conservation programs northern fulmars. Ecology 77, 1181 1188. Furness, R.W., Todd, C.M., 1984. Diets and feeding of fulmars Fulmarus glacialis that focus entirely on territorial waters adjacent to breeding sites during the breeding season: a comparison between St Kilda and Shetland are likely to underestimate the management practices that are colonies. Ibis 126, 379–387. necessary to wide-ranging species such as fulmars. Garthe, S., Furness, R.W., 2001. Frequent shallow diving by a northern fulmar feeding at shetland. Waterbirds: the International Journal of Waterbird Biology 24, 287–289. Grosbois, V., Thompson, P.M., 2005. North Atlantic climate variation influences Acknowledgements survival in adult fulmars. Oikos 109, 273–290. Hamer, K.C., Thompson, D.R., 1997. Provisioning and growth rates of nestling fulmars Fulmarus glacialis: stochastic variation or regulation? Ibis 139, 31–39. We thank Orkney Islands Council for access to Eynhallow and Hatch, S.A., Gill, V.A., Mulcahy, D.M., 2010. Individual and colony-specific wintering Talisman Energy (UK) Ltd and Marine Scotland for fieldwork and areas of Pacific northern fulmars (Fulmarus glacialis). Can. J. Fish. Aquat. Sci. 67, equipment support. Handling and tagging of fulmars was con- 386–400. Letessier, T.B., Falkenhaug, T., Debes, H., Bergstad, O.A., Brierley, A.S., 2011. ducted under licences from the British Trust for Ornithology and Abundance patterns and species assemblages of euphausiids associated with the UK Home Office. Microwave SST data were provided by RSS the Mid-Atlantic Ridge, North Atlantic. J. Plankton Res. 33, 1510–1525. (Remote Sensing Systems, http://www.ssmi.com), and processed Letessier, T.B., Pond, D.W., McGill, R.A.R., Reid, W.D.K., Brierley, A.S., 2012. Trophic by the NERC Earth Observation Data Acquisition and Analysis interaction of invertebrate zooplankton on either side of the Charlie Gibbs Fracture Zone/Subpolar Front of the Mid-Atlantic Ridge. J. Mar. Syst. 94, Centre, Plymouth. Ewan Edwards was supported by a Marine 174–184. Alliance for Science and Technology Scotland studentship, and Mackley, E., Phillips, R., Silk, J., Wakefield, E., Afanasyev, V., Furness, R., 2011. At-sea activity patterns of breeding and nonbreeding white-chinned petrels Procellaria Lucy Quinn was supported by a NERC Studentship. Thanks also to – fi aequinoctialis from South Georgia. Mar. Biol. 158, 429 438. the many colleagues who assisted with eldwork during the Magalhães, M.C., Santos, R.S., Hamer, K.C., 2008. Dual-foraging of Cory's shear- project, and two anonymous reviewers for their comments on waters in the Azores: feeding locations, behaviour at sea and implications for the manuscript. food provisioning of chicks. Mar. Ecol. Prog. Ser. 359, 283–293. Mallory, M., Akearok, J., Edwards, D., O‘Donovan, K., Gilbert, C., 2008a. Autumn migration and wintering of northern fulmars (Fulmarus glacialis) from the Canadian high Arctic. Polar Biol. 31, 745–750. Appendix A. Supplementary material Mallory, M., Gaston, A., Forbes, M., Gilchrist, H., Cheney, B., Lewis, S., Thompson, P., 2008b. Flexible incubation rhythm in northern fulmars: a comparison between – Supplementary material related to this article can be found oceanographic zones. Mar. Biol. 154, 1031 1040. Miller, P., 2009. Composite front maps for improved visibility of dynamic sea- online at http://dx.doi.org/10.1016/j.dsr2.2013.04.011. surface features on cloudy SeaWiFS and AVHRR data. J. Mar. Syst. 78, 327–336. Miller, P.I., Read, J.F., Dale, A.C., 2013. Thermal front variability along the North References Atlantic Current observed using satellite microwave and infrared data. Deep Sea Res. Part II: Top. Stud. Oceanogr. 98 (PB), 244–256. Mitchell, P.I., Newton, S.F., Ratcliffe, N., Dunn, T.E., 2004. Seabird populations of Anderson, O.R.J., Small, C.J., Croxall, J.P., Dunn, E.K., Sullivan, B.J., Yates, O., Black, A., Britain and Ireland. Poyser, London. 2011. Global seabird bycatch in longline fisheries. Endangered Species Res. 14, OSPAR, 2010. Background Document on the Charlie-Gibbs Fracture Zone. OSPAR 91–106. Publication 2010/523. Boertmann, D., 2011. Seabirds in the central North Atlantic, September 2006: Owen, E., Daunt, F., Moffat, C., Elston, D., Wanless, S., Thompson, P.M. Analysis of fatty further evidence for an oceanic seabird aggregation area. Mar. Ornithol. 39, acids and fatty alcohols reveals seasonal and sex-specific changes in the diets of 183–188. seabirds. Mar. Biol., http://dx.doi.org/10.1007/s00227-012-2152-x,inpress. Bogdanova, M.I., Daunt, F., Newell, M., Phillips, R.A., Harris, M.P., Wanless, S., 2011. Pennycuick, C.J., 2002. Gust soaring as a basis for the flight of petrels and Seasonal interactions in the black-legged kittiwake, Rissa tridactyla: links albatrosses. Avian Sci. 2, 1–12. between breeding performance and winter distribution. Proc. R. Soc. B: Biol. Phillips, R.A., Hamer, K.C., 2000. Growth and provisioning strategies of Northern Sci. 278, 2412–2418. Fulmars Fulmarus glacialis. Ibis 142, 435–445. Calenge, C., 2006. The package adehabitat for the R software: a tool for the analysis Phillips, R.A., Petersen, M.K., Lilliendhal, K., Solmundsson, J., Hamer, K.C., Cam- of space and habitat use by animals. Ecol. Modelling 197, 516–519. phuysen, C.J., Zonfrillo, B., 1999. Diet of the northern fulmar Fulmarus glacialis: Doksæter, L., Olsen, E., Nøttestad, L., Fernö, A., 2008. Distribution and feeding reliance on commercial fisheries? Mar. Biol. 135, 159–170. ecology of dolphins along the Mid-Atlantic Ridge between Iceland and the Phillips, R.A., Silk, J.R.D., Croxall, J.P., Afanasyev, V., Briggs, D.R., 2004. Accuracy of Azores. Deep Sea Res. Part II: Top. Stud. Oceanogr. 55, 243–253. geolocation estimates for flying seabirds. Mar. Ecol. Prog. Ser. 266, 265–272. Dunnet, G.M., 1991. Population studies of the Fulmar on Eynhallow, Orkney Islands. Pinaud, D., Weimerskirch, H., 2007. At-sea distribution and scale-dependent Ibis 133, 24–27. foraging behaviour of petrels and albatrosses: a comparative study. J. Anim. Dunnet, G.M., Anderson, A., 1961. A method for sexing living Fulmars in the hand. Ecol. 76, 9–19. Bird Study 8, 119–126. Pirazzoli, P., Tomasin, A., Ullmann, A., 2010. Recent changes in measured wind in Egevang, C., Stenhouse, I.J., Phillips, R.A., Petersen, A., Fox, J.W., Silk, J.R.D., 2010. the NE Atlantic and variability of correlation with NAO. AnnalesGeophysicae 28, Tracking of Arctic terns Sterna paradisaea reveals longest animal migration. 1923–1934. Proc. Natl. Acad. Sci. 107, 2078–2081. R Development Core Team, 2008. R: A Language and Environment for Statistical European Commission, 2009. Directive 2009/147/EC of the European Parliament Computing. Vienna, Austria. and the Council of 30 November 2009 on the conservation of wild birds. Off. Skov, H., Gunnlaugsson, T., Budgell, W.P., Horne, J., Nøttestad, L., Olsen, E., Søiland, J. Eur. Communities L 20, 7–25. H., Víkingsson, G., Waring, G., 2008. Small-scale spatial variability of sperm and Falk, K., Moeller, S., 1997. Breeding ecology of the Fulmar Fulmarus glacialis and the sei whales in relation to oceanographic and topographic features along the Kittiwake Rissa tridactyla in high-arctic northeastern Greenland, 1993. Ibis 139, Mid-Atlantic Ridge. Deep Sea Res. Part II: Top. Stud. Oceanogr. 55, 254–268. 270–281. Sutton, T.T., Porteiro, F.M., Heino, M., Byrkjedal, I., Langhelle, G., Anderson, C.I.H., FAO, 2008. Report of the Expert Consultation on Best Practice Technical Guidelines Horne, J., Søiland, H., Falkenhaug, T., Godø, O.R., Bergstad, O.A., 2008. Vertical for IPOA/NPOA–Seabirds. Bergen Norway, 2–5 September 2008 (No. 880). FAO, structure, biomass and topographic association of deep-pelagic fishes in Rome. relation to a mid-ocean ridge system. Deep Sea Res. Part II: Top. Stud. Oceanogr. Fauchald, P., Tveraa, T., 2003. Using first-passage time in the analysis of area- 55, 161–184. restricted search and habitat selection. Ecology 84, 282–288. Thaxter, C.B., Lascelles, B., Sugar, K., Cook, A.S.C.P., Roos, S., Bolton, M., Langston, R. Fossen, I., Cotton, C.F., Bergstad, O.A., Dyb, J.E., 2008. Species composition and H.W., Burton, N.H.K., 2012. Seabird foraging ranges as a preliminary tool for distribution patterns of fishes captured by longlines on the Mid-Atlantic Ridge. identifying candidate Marine Protected Areas. Biol. Conserv. 156, 53–61. Deep Sea Res. Part II: Top. Stud. Oceanogr. 55, 203–217. Thompson, P.M., Ollason, J.C., 2001. Lagged effects of ocean climate change on Fox, J.W., 2010. Geolocator Manual v8 (March 2010). fulmar population dynamics. Nature 413, 417–420. Frederiksen, M., Moe, B., Daunt, F., Phillips, R.A., Barrett, R.T., Bogdanova, M.I., Tickell, W.L.N., 2000. Albatrosses. Pica Press, Sussex, UK. Boulinier, T., Chardine, J.W., Chastel, O., Chivers, L.S., Christensen-Dalsgaard, S., Weimerskirch, H., Louzao, M., De Grissac, S., Delord, K., 2012. Changes in wind Clément-Chastel, C., Colhoun, K., Freeman, R., Gaston, A.J., González-Solís, J., pattern alter albatross distribution and life-history traits. Science 335, 211–214.